Thread: dhcpd.conf not configured after installation

Hello,

I downloaded the following KUSU image: kusu-1.1.centos-5.x86_64.iso

I used CentOS-5.2-x86_64-bin-DVD.iso to create an 'OS' kit.

I did not have any problems during installation, however when I tried to provision a node, I noticed that DHCP server is not running on the installer node.

The reason is that dhcpd.con file is empty. When I tried re-creating it like this:

Code:

 # genconfig dhcpd > /etc/dhcpd.conf

I get the following traceback:

Code:

[root@benvmkusu11 ~]# genconfig dhcpd
Traceback (most recent call last):
  File "/opt/kusu/bin/genconfig", line 133, in ?
    app.run()
  File "/opt/kusu/bin/genconfig", line 127, in run
    dbrinst.run(self.plugin, self.action, self.pluginargs)
  File "/opt/kusu/lib/python/kusu/genconfig.py", line 77, in run
    pinst.runPlugin(pargs)
  File "/opt/kusu/lib/plugins/genconfig/dhcpd.py", line 173, in runPlugin
    [x.ip for x in network.hosts])
  File "/opt/kusu/lib/plugins/genconfig/dhcpd.py", line 212, in getNetworkRanges
    index = ip_range.index(h)
ValueError: list.index(x): x not in list
[root@benvmkusu11 ~]#

Here is the network configuration on the cluster:

Code:

[root@benvmkusu11 ~]# netedit -l

Network ID Device       Network       Subnet          Network Description                      Interface Type
========== ======       =======       ======          ===================                      ==============
1          eth1         172.25.254.0  255.255.255.0   public                                   public
2          eth0         10.10.6.0     255.255.255.0   cluster                                  provision

Any suggestion?

Hi

Can you provide us with the network information of both eth0 and eth1 nics via ifconfig -a?

We may have already fixed this bug. Can you replace /opt/kusu/lib/plugins/genconfig with http://svn.osgdc.org/kusu-repo/kusu/...onfig/dhcpd.py. You will need to restart dhcpd after running genconfig dhpcpd > /etc/dhcpd.conf

-Liming

Hi Liming,

Here is the output of 'ifconfig -a':

Code:

[root@benvmkusu11 ~]# ifconfig -a
eth0      Link encap:Ethernet  HWaddr 00:0C:29:EB:75:69
          inet addr:10.10.6.1  Bcast:10.10.6.255  Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:feeb:7569/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:8 errors:0 dropped:0 overruns:0 frame:0
          TX packets:54 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:4720 (4.6 KiB)  TX bytes:9437 (9.2 KiB)
          Base address:0x2000 Memory:d8920000-d8940000

eth1      Link encap:Ethernet  HWaddr 00:0C:29:EB:75:73
          inet addr:172.25.254.76  Bcast:172.25.254.255  Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:feeb:7573/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:9067 errors:0 dropped:0 overruns:0 frame:0
          TX packets:1257 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:818856 (799.6 KiB)  TX bytes:131002 (127.9 KiB)
          Base address:0x2040 Memory:d8940000-d8960000

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:983 errors:0 dropped:0 overruns:0 frame:0
          TX packets:983 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:1085584 (1.0 MiB)  TX bytes:1085584 (1.0 MiB)

sit0      Link encap:IPv6-in-IPv4
          NOARP  MTU:1480  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)

[root@benvmkusu11 ~]#

I will do as you suggest and try provisioning again.

Thanks,
Uncas

Hi Liming,

Using the patched dhcpd.py file did the trick. After replacing the original file, I was able to regenerate the dhcpd.conf file and restart the DHCP server. I provisioned one node without problems.

Thanks for quick reply,
Uncas

Hi

Not a problem. Do let us know if you have encountered any problems.

-Liming

I too lost my dhcpd.conf information on my front end node. I've tried the genconfig command as seen above and it fails the same way. I downloaded and replaced (backing up the original) the /opt/kusu/lib/plugins/genconfig/dhcpd.py and it did not fix my problem.

Here is the output from the genconfig dhcpd > /etc/dhcpd.conf
[root@icr-eth1 genconfig]# genconfig dhcpd > /etc/dhcpd.conf
Traceback (most recent call last):
File "/opt/kusu/bin/genconfig", line 133, in ?
app.run()
File "/opt/kusu/bin/genconfig", line 127, in run
dbrinst.run(self.plugin, self.action, self.pluginargs)
File "/opt/kusu/lib/python/kusu/genconfig.py", line 77, in run
pinst.runPlugin(pargs)
File "/opt/kusu/lib/plugins/genconfig/dhcpd.py", line 146, in runPlugin
[x.ip for x in network.hosts])
File "/opt/kusu/lib/plugins/genconfig/dhcpd.py", line 174, in getNetworkRanges
network = IP('%s/%s' % (subnet, netmask))
File "/usr/lib/python2.4/site-packages/IPy.py", line 203, in __init__
raise ValueError, "%s has invalid prefix length (%s)" % (repr(self), self._prefixlen)
ValueError: IP('192.168.0.1/24') has invalid prefix length (24)


That network (192.168.0.1/24) would be my infiniband IPoIB network. I need dhcp to service my private network (provision network) of 172.33.0.0 which is listed as the 1st of 3 networks in netedit -l. (private, public, Infiniband Net)

Any other ideas?

Hi,

Can you set your infiniband network to be 192.168.0.0/24 instead? The DHCPD plugin uses the IPy package, which cannot generate an IP range for the values given.